Algorithm

Quantum Fourier Transform

Maps the computational basis to the Fourier (frequency) basis. Subroutine of QPE and Shor.

Qubits 3 Cost O(n²) Default 3 qubits, input |000⟩

On n qubits, QFT|x⟩ = 2⁻ⁿ/² Σ_k exp(2πi x k / 2ⁿ) |k⟩. The circuit is H on each wire from MSB to LSB, with controlled phases CP(π/2^{k−1}) from lower wires, then SWAP to reverse wire order.

Qubi CP angles are π-multiples: CP [j,i] 0.5 is a π/2 controlled phase. Applied to |000⟩ the QFT is a uniform superposition.

How it works

  1. 1
    MSB Hadamard and phases

    H 2, then CP from q1 (angle 1/2) and q0 (angle 1/4) onto q2.

  2. 2
    Middle wire

    H 1, then CP [0,1] 0.5.

  3. 3
    LSB and reverse

    H 0, then SWAP [0,2] so the Fourier bit order matches the usual convention.

Default Qubi (static)
H 2
CP [1,2] 0.5
CP [0,2] 0.25
H 1
CP [0,1] 0.5
H 0
SWAP [0,2]

Function form: QFT(0..2)

Circuit
q0q1q2HP0.5P0.25HP0.5H
Expected result

State. QFT|000⟩ = |+⟩⊗³ = 2⁻³/² Σ_x |x⟩

Each of the 8 computational basis states has probability 1/8.

BasisAmplitudeProbability
|000⟩ ... |111⟩ 1/√8 each
12.5%

Notes

  • Classical FFT is O(n 2ⁿ) on the same 2ⁿ amplitudes. QFT is O(n²) gates.
  • IQFT is the exact adjoint (see Inverse QFT).

Change parameters

The write-up above is for the default circuit. Use this control to generate other variants and load them in the simulator.